1

Spectroscopic Study of the Thermal Degradation of PVP-capped Rh

and Pt Nanoparticles in H2 and O2 environments

Yuri Borodko,a Hyun Sook Lee,a Sang Hoon Joo,b Yawen Zhang,b and Gabor Somorjai a,b,*

a) Chemical and Materials Sciences Divisions, Lawrence Berkeley National Laboratory, 1 Cyclotron Road, Berkeley CA 94720

b) Department of Chemistry, University of California, Berkeley, CA 94720

Abstract

Poly(N-vinylpyrrolidone (PVP) capped platinum and rhodium nanoparticles (7 - 12

nm) have been studied with UV-VIS, FTIR and Raman spectroscopy. The absorption

bands in the region 190 – 900 nm are shown to be sensitive to the electronic structure of

surface Rh and Pt atoms as well as to the aggregation of the nanoparticles.

In-situ FTIR-DRIFT spectroscopy of the thermal decay of PVP stabilized Rh and Pt

nanoparticles in H2 and O2 atmospheres in temperatures ranging from 30 oC – 350 oC

reveal that decomposition of PVP above 200 oC, PVP transforms into a ‘polyamid-

polyene’ - like material that is in turn converted into a thin layer of amorphous carbon

above 300 oC. Adsorbed carbon monoxide was used as a probing molecule to monitor

changes of electronic structure of surface Rh and Pt atoms and accessible surface area.

The behavior of surface Rh and Pt atoms with ligated CO and amide groups of

pyrrolidones resemble that of surface coordination compounds.

Introduction Poly(vinylpyrrolidone) (PVP) is used extensively as a flexible membrane for stabilizing

metallic nanocrystals with varying size and shape [1]. The application of Pt-PVP and Rh-

PVP nanoparticles in catalytic reactions has recently offered promising new opportunities

to create tailored catalytic systems [2]. The knowledge of the structure and properties of

the interface between the capping agent and the metal surface is critical for understanding

nanoparticles role in catalytic reactions, since capping agents can stabilize specific

2

electronic states of the surface atoms, thus creating specific catalytic active sites. A

routine procedure used to activate heterogeneous catalysts is redox cycling treatment of

the catalysts at elevated temperatures. This may induce changes in the electronic structure

of metallic atoms, the size and shape of the nanoparticles, the chemical state of the

capping agent, and in the area of metallic surface that is accessible to the reactants. To

minimize temperature effects and retain the shape and size of Pt-PVP and Rh-PVP

nanoparticles, temperatures usually do not exceed 100º-300 ºC. However, in this

temperature range capping agents can be decomposed, especially in reactive media such

as O2 and H2. It should be noted, that treatment of organic stabilizer in ozone-reach

atmosphere is promising option for removing the capping agent at low temperature [2a].

In this paper, we report a study of the structure and stability of pure PVP and PVP-

capped Rh and Pt nanoparticles using in-situ spectroscopic measurements. UV-VIS

spectra reveal that the charge-transfer (CT) band (330 nm) and optical surface plasma-

resonance absorption (245 nm), are surface sensitive. The intensity of the CT band

shows sensitivity towards the electronic structure of Pt and Rh atoms on the surface as

well as the aggregation of the nanoparticles, whereas optical plasmon band appears only

for aggregated nanoparticles. FTIR-DRIFT studies as a function of temperature found

that the decay of the inner PVP capping layer, which is bonded to the surface metal

atoms, begins ~100 ºC lower than the decay of the outer unbound PVP capping layer or

of pure PVP. Carbon monoxide chemisorbed on the surface of metal nanoparticles shows

subtle IR changes that are a result of donor-acceptor interactions between polymer amide

groups and surface metal atoms. Formation of some intermediates that form during the

degradation of capping PVP on Pt and Rh as the temperature is increased was detected in

3

the IR with characteristic bands: NH (3342 cm-1), CH (3089 cm-1), C≡N (2246 cm-1) and

C=C-C=C (1595 cm-1). Further heating of Rh- and Pt-PVP above 300 ºC leads to the

formation of a thin layer of a complex composite: ‘polyamide-polyene’ like fragments

together with amorphous carbon.

Experimental Section

Chemicals. Rhodium (III) acetylacetonate (Rh(acac)3, 97%), Platinum (II)

acetylacetonate (Pt(acac)2, 97%), 1,4-butandiol(99%), poly(vinylpyrrolidone) (PVP,

Mw=55K), and acetylacetone (Hacca, 99%) were purchased from Aldrich and used as

received. All solvents (acetone, ethanol, hexane and chloroform) were analytical grade

and used without further purification.

Synthesis of Pt and Rh nanoparticles. For synthesis of PVP capped Rh

nanoparticles we use 0.25 mmol Rh(acac)3 with 2.5 mmol of PVP dissolved in 20 ml of

1,4-butanediol [3]. The stock solution was heated to 140 ºC in a Glass-Col electromantle

(60W; 50ml) and was evacuated at this temperature for 20min to remove water and

oxygen under magnetic stirring, resulting in an optically transparent orange-yellow

solution. The flask was then heated to the desired reaction temperature, between 170 and

230 oC, at a rate of 10 oC min-1, and maintained at this temperature (±2 oC) for 2 hours

under an Ar atmosphere. During the reaction, the color of the solution gradually turned

from orange-yellow to black. When the reaction was complete, an excess of acetone was

poured into the solution at room temperature to form a cloudy black suspension. This

suspension was separated by centrifugation at 4200 rpm for 6 min, and the black product

was collected by discarding the colorless supernatant. The precipitated Rh nanoparticles

were washed with acetone once and then redispersed in ethanol in three-necked flask at

4

room temperature. Platinum nanoparticles were prepared by using the same method, but

instead using 0.1 mmol Pt(acac)2 with 1 mmol PVP dissolved in 20 ml of 1,4-butanediol

as starting materials. TEM analysis of the Rh or Pt nanoparticles analysis was carried out

with a Philips FEI Tecnai 12 transmission electron microscope operated at 100 kV. The

average size of Pt nanoparticles was 12 ± 3 nm and 8 ± 2 nm for Rh nanoparticles (Figure

1).

Spectroscopic Characterization

UV-VIS. Optical properties of Pt and Rh nanoparticles were characterized using a UV-

VIS spectrophotometer, Perkin-Elmer, LAMDA 650 (190 nm-900 nm). Spectra were

taken of drop casted films deposited on a quartz window in transmission or Al-foil in

reflectance with an integrating sphere of 60 mm in diameter coated with spectralon .

The morphology of Rh- and Pt-PVP nanoparticles deposited on Si-wafers was

characterized with SEM using a Zeiss Gemini Ultra-55 analytical scanning electron

microscope, before and after heating in air.

FTIR-DRIFT spectra were measured with a Nicolet Nexus-670 spectrophotometer with

integrated diffuse reflectance optics (Spectra-Tech Collector II). The improved spectra

could be obtained by using a thin layer of PVP/Pt or PVP/Rh deposited on aluminum foil.

Since the layer thickness can be adjusted so as to render the sample partially IR

transparent, we were able to optimize the ratio of diffuse to specular reflectance and

obtain spectra of similar quality compared to those measured in transmission mode [4].

The temperature/vacuum chamber Spectra-Tech 0030-101 was used for in-situ FTIR

measurements in 5% H2 balanced with N2 and pure O2 flow, or a static CO atmosphere.

An oil-free Alcatel Drytel 34 turbo-pump station was used for sample evacuation.

5

The deep UV-Raman spectroscopy system was described in detail elsewhere [5,

6]. A continuous wave (cw) intracavity double argon ion laser operating at 244 nm is

used as the excitation source. Backscattered light was collected and directed to the triple

spectrometer Spex 1877C, which was optimized for performance in the deep UV region.

Spectra were recorded with an LN2-cooled, UV-enhanced CCD camera. The instrument

dispersion is 2.1 cm-1/pixel; the typical resolution was 12 cm-1.

To eliminate decomposition of Pt and Rh-PVP samples under UV irradiation we used a

custom designed rotating sample holder (400-1200 rpm) equipped with a second motor

for lateral translation (240 rpm). In this manner, the laser beam was XY-rastered over the

entire sample (deposited on Al-foil) decreasing the UV exposure in a given spot by 10-3

when compared with a static sample holder. Typical collection times were 5-30 min,

using 3-5 mW of 244 nm excitation energy. GRAMS/AI and Omnic software from

Thermo Galactic was used for processing FTIR-DRIFT and UV-Raman spectra.

X-ray diffraction patterns were measured on a Bruker D8 GADDS diffractometer using

Co Kα radiation (1.79 Å). XRD samples were prepared by depositing the nanoparticles

on a glass plate. Different chemically-inert supporting materials were used for

characterization of the metal nanoparticles: quartz windows for UV-VIS transmission;

Al-foils for UV-VIS reflection and FTIR-DRIFT; glass plates for XRD and Si-wafers for

SEM.

RESULTS and DISCUSSION

UV-VIS spectra of Rh-PVP and Pt-PVP

Recently, UV-VIS absorbance spectra for Rh-PVP nanoparticles in solution have been

presented in [7]. We have taken the UV-VIS spectra of PVP, Rh-PVP, and Pt-PVP as: i)

6

absorbance of a homogeneous solution, ii) absorbance of a drop-cast film deposited on a

quartz window, and iii) reflectance of a film deposited on aluminum foil in the region of

200-900 nm. The UV-VIS spectrum of PVP has a narrow intense band at 200 nm, which

may be assigned to n � ππππ* transitions between the oxygen lone pairs and the vacant ππππ*-

orbital of the pyrrolidone ring (Fig. 2A). The absorbance spectra of PVP are very similar

for cases i and ii. In the spectra of Rh-PVP and Pt-PVP, along with the n �ππππ* band at

200 nm, new bands appear near 330 nm and 300 nm respectively, which may be

assigned to charge-transfer transitions of an electron excited from the carbonyl lone pair

to a vacant d-orbital of the metallic nanoparticles (Fig. 2B). Thus, the spectra of Rh-PVP

and Pt-PVP nanoparticles in the UV region have two bands, one being characteristic of a

loose shell of the polymeric stabilizer around a metallic core and the other of a single

surface layer formed between chemisorbed carbonyl groups and surface metallic atoms

>C=O�Rhδδδδ+ or Ptδδδδ+. This interpretation of UV absorption spectra of Rh-PVP and Pt-

PVP is in good agreement with our recent observations of CT-SERS effects for these

nanoparticles by deep UV Raman [5, 6]. It was found that the UV-VIS spectra of

nanoparticles differ substantially depending on the sample aggregation. The aggregation

causes multiple scattering and changes the spectra depending on the mode of

measurement (transmission or reflection) [8]. As seen in Figures 2B and 2C, the ratio of

n→→→→ ππππ* band to that of the CT band for a homogeneous EtOH solution is higher in

transmission than for a drop-cast film deposited on Al-foil in reflection. This is a result

of multiple scattering of light on aggregated Rh-PVP nanoparticles. The thicker the layer

of Rh-PVP on Al foil, the stronger is the CT band at 330 nm relative to the intensity of

the n→→→→ ππππ* band at 200 nm (Figure 2C). The contribution of multiple reflections from the

7

surface of metallic nanoparticles in thick films increases due to the aggregation of

nanoparticles, leading to substantial increases in CT band intensity. The thickness of

drop-cast layers was estimated from the Fabry-Perot fringe separation that arise from the

interference between beams transmitted and partially reflected at the film-quartz interface

shown in the insert in Figure 2D. The effective thickness of drop-cast films under study

was in the range 1.5-5µm , calculated using the expression d = ∆m/[2(n2 – Sin

2Θ)

1/2∆νi

f], where ∆m is the number of fringes, ∆νif is the wavenumber difference between the

initial and final fringe, and Θ is the angle of incidence [9]. The value of the refractive

index n for Pt-PVP and Rh-PVP layers was taken as 1.5.

Influence of temperature on UV-VIS spectra. Rh and Pt metallic nanoparticles

stabilized by PVP are characterized by four types of absorption bands in the UV-VIS

region (Figure 3). Three types of electronic transition are inherently characteristic of

nanoparticles when the size is less than the mean free path of electrons. The additional

band for optical plasma-resonance absorption should develop as a result of aggregation of

metallic nanoparticles when the aggregate size is above 20-30 nm. UV-VIS reflection

spectra of Pt and Rh nanoparticles subjected to reduction in hydrogen flow at

temperatures ranging from 30°-170°C show a weak dependence on the temperature

(Figure 3A). At 170°C the intensity of charge-transfer bands slightly decreased due to

reduction of Pt/Rh atoms on the surface of nanoparticles, since the interaction of a

carbonyl group of PVP and Pt/Rh atoms on oxidized metal nanoparticle surfaces is

stronger than on the reduced surfaces [5, 6]. UV-VIS spectra were also studied for drop-

casted samples deposited on a quartz window and subjected to heating in air in the

temperature range 20°-550°C. Recently it was shown that during the oxidation of Rh

8

nanoparticles in air at elevated temperatures, both the decomposition of the PVP layer

and the oxidation of rhodium atoms on the surface to form a RhOx layer occur [10]. Pt

nanoparticles (with diameters > 3 nm) are more stable towards oxidation and produce

only a single oxide layer of PtOx [11]. As can be seen in Figure 3B, after treatment in air

at elevated temperatures, the band at 200 nm (n→→→→ ππππ* for loosely bonded pyrrolidone

rings) and CT bands at 330 nm disappear and new bands at 227 nm, 370 nm, and 590 nm

gradually appear. This transformation of UV spectrum suggests decomposition of the

capping PVP and aggregation of metallic nanoparticles. With this possibility in mind, the

X-ray diffraction for nanoparticles heated in air was studied.

The XRD diffraction patterns of Pt and Rh nanoparticles are shown in Fig 4 for samples

heated in air at 50 °C, 250 °C, and 450 °C for 15 min. The three peaks correspond to

Bragg reflections of the (111), (200), and (220) crystal planes of the metallic core of Pt

and Rh nanoparticles. The inserts in Figures 4A and 4B reveal that the (111) reflection

becomes sharper after heating at higher temperatures. The mean diameter of the

nanoparticles (calculated by the Debye-Scherrer formula) increases about 20% and 35%

for Pt and Rh, respectively, after heating to 450 °C (Table 1).

TABLE 1

SAMPLE FWHM ΘΘΘΘ d(nm)

Pt/PVP_50 °C 1.02 23.32 9.7

Pt/PVP_250 °C 0.92 23.32 10.8

Pt/PVP_450 °C 0.85 23.27 11.7

Rh/PVP_50 °C 1.83 24.00 5.5

Rh/PVP_250 °C 1.85 24.00 5.4

Rh/PVP_450 °C 1.35 24.06 7.4

9

SEM images show that the morphology of Rh-PVP and Pt-PVP drop-cast films deposited

on Si wafers changes substantially after heating. At temperatures above 400 °C, the 9.7

nm Pt and 5.5 nm Rh nanoparticles form 50-80 nm aggregates (Figure 4). To gain insight

into how aggregation changes the optical properties of the nanoparticles, backward

scattered light of Rh-PVP deposited on a quartz window was measured using an

integrating sphere. In this experimental configuration, the sample is placed on the port of

the integrating sphere and a black absorber behind the sample is used to minimize the

contribution of specular reflection and transmitted light (Figure 5A, 5B). The reflectance

spectrum is shown in Figure 5C. The pronounced absorption peak at 244 nm may be

assigned to the optical plasma-resonance absorption of Rh-aggregates. Recently the

optical properties of metallic Rh [12] and metallic-island Rh films [13] were studied.

Features were observed at 1.1 eV (1128 nm), 2.68 eV (463 nm) and 5.8 eV (214 nm) in

reflection spectra of the polycrystalline samples, which were interpreted as interband

transitions [12]. The transmittance spectra of Rh island films with different particle

diameters in the range of 55 – 280 nm exhibit bands at 2.15 eV (577 nm), 3.5 eV (355

nm) and 6.1 eV (203.5 nm). The dominant band at 3.5 eV (diameter ~ 280 nm) was

assigned to the optical plasma-resonance absorption [13]. The position of UV-VIS bands

for Rh island films and for Rh nanoparticle aggregates that formed at high temperature in

air are surprisingly close, even though the surface of Rh aggregates is covered by

rhodium oxide. It is worth noting that 38 nm diameter Pt and Pd particles are

characterized by optical plasmon resonance absorptions [14] at the same wavelength as

those of the plasmon band for Rh aggregates. The effective size of our Rh nanoparticle

aggregates was estimated from the optical plasma - resonance absorption using the

10

dependence of optical plasma resonance on the size of Rh islands as determined by Anno

[13] (Fig 6a). The absorption at 244 nm corresponds to an effective size of ~ 30 nm. This

value is substantially higher than the 7.4 nm obtained by XRD measurements. In XRD,

diffraction occurs on crystalline metallic nanoparticles and peaks become broader if there

are fewer planes that give the rise to Bragg reflection. The increase in effective diameter

from 5.5 to 7.4 nm may be interpreted as two Rh nanoparticles merging together (Figure

6b). The average size of Rh nanoparticles estimated from the location of the optical

surface plasma- resonance [15a] reflects the size of electronically-connected clusters of

metallic nanoparticles . This size is substantially higher than the size from X-ray

scattering measurements, which give the size of the discrete Rh nanocrystals based on the

coherence length of the atoms in the merged nanoparticles. The effect of nanoparticle

aggregation on optical plasma-resonance position is due to organization of isometric

metallic nanoparticles into arrays/clusters, which exhibit strong coupling to surface

plasmons (red-shift of the SPR band) while leaving discreet metallic nanoparticles intact

[15b].

Thermal Decomposition of PVP, Pt-PVP and Rh-PVP.

Conformational change of PVP. Decay in H2, N2 and O2. The thermal stability of pure

PVP has been extensively studied [16]. However, considerably less is known about the

transformation of PVP capping agent on the surface of Rh and Pt nanocrystals subjected

to high-temperature treatment under reducing and oxidizing conditions, a standard

procedure in heterogeneous catalyst preparation. Monitoring the in-situ transformation of

PVP, Rh-PVP, and Pt-PVP in the temperature range of 25-350 °C (heating rate

3.2°C/min) in N2, H2 and O2 flow was done using a DRIFTS reactor. Assignments of

11

FTIR and Raman spectra were presented in a previous work [6]. The heating of PVP in

H2 and N2 flow up to 200 °C did not substantially change the position of carbonyl band at

1680 cm-1. Very small changes are seen in the region of CH stretching bands assigned to

vibrations of the CH2 groups in the chain and ring of PVP. As seen in Figure 7a, the band

at 2873 cm-1 (assigned to tertiary C-H stretching vibrations) is more sensitive to the

temperature changes. Substantial changes are also seen in bands at 1430 cm-1 (CH2

scissors) and at 1280 cm-1 (CH2 wag.) (Figure 7b). It should also be noted that the

presence of isosbestic points in the spectra (Figure 7a, b) indicates the initial form of PVP

converted into a new one quantitatively. This data shows that, up to 200 °C, PVP does

not decompose in an H2 atmosphere, but that the PVP chain structure and packing density

changes, most likely due to a glass-transition and cross-linking [16]. Degradation of PVP

in O2 atmosphere starts at about 100 °C. The maximum of the carbonyl band shifts from

1680 cm-1 at 34 °C to 1720 cm-1 at 190 °C, a possible indication of the formation of

aliphatic ketones [17]. The C-H stretching and bending modes of PVP simultaneously

decrease in intensity above 100 °C, while two new bands and a doublet appear at 3325

cm-1, 3075 cm-1, and 2250-2223 cm-1 respectively.

Decay of capping PVP starts at the interface of PVP/Pt.

The decomposition of Pt-PVP and Rh-PVP nanoparticles was studied under similar

experimental conditions. The spectra of PVP capping agent in N2 and H2 atmospheres

gradually decrease in intensity above 200 °C. However, the various spectra for the initial

Rh-PVP and its products in the temperature range of 30-200°C behave similarly, whereas

in the case of Pt-PVP, new absorption bands appear near 2000 and 1810 cm-1 in the

temperature range of 85-215 °C (Figure 7c). The appearance of the new bands

12

characteristic of linear and bridging carbon monoxide, indicates that pyrrolidone rings

adsorbed on platinum atoms begin to decompose at ~100°C lower than the decomposition

temperature for non-bonded amide groups as determined in present work by FTIR and by

TGA [16]. It is interesting to note that Pt and Rh nanoparticles act as a trap for the carbon

monoxide that is eliminated from side pyrrolidone rings adjacent to the metallic surface

during degradation. In the H2 atmosphere, loosely bonded PVP chains around Pt and Rh

nanoparticles maintain their composition until 200 °C, in contrast to a thin layer of

pyrrolidone rings adjacent to metal surface. The difference in thermal decomposition of

the interfacial layer for Rh-PVP and Pt-PVP can be seen from a comparison of the

intensity ratio for characteristic bands of amide carbonyls at 1680 cm-1 and C-H

deformation modes at 1430 cm-1 and 1370 cm-1 for residual PVP (Figure 7c and 10c).

The ratio of intensities C=O/C-H did not change in the case of Rh-PVP up to 220 °C;

however the shape of C-H deformation modes change due to packing and cross-linking of

the PVP capping agent. In the case of Pt-PVP, more noticeable features appear: the

intensities of CH bands substantially decrease relative to the C=O carbonyl stretch and

new carbon monoxide bands at 2020 cm-1 and a doublet centered at 1810 cm-1 appear in

the temperature range of 80 -180 °C (Figure 7c).

Carbon monoxide adsorbed on Rh-PVP and Pt-PVP monitors the

electronic state of metallic surface during PVP degradation.

Both the frequency and the intensity for the C≡O stretch of carbon monoxide adsorbed on

Pt-PVP in a H2 flow show (in-situ measurement) unusual maxima as functions of

temperature. The frequencies for both linear and bridging carbon monoxides increase up

to 160 °C, and decreases thereafter (Figure 8a). Under most circumstances, the frequency

13

of the carbon monoxide decreases if the surface of Pt nanoparticles is reduced in

hydrogen at higher temperatures, since the ability of surface Pt atoms to contribute in

dπ→π* back-bonding to adsorbed carbon monoxide is increased. It was previously

shown that on the surface of Pt nanoparticles above 2 nm in size, only a monolayer of

PtO, which is reducible by H2 at 50-100 °C [11], was formed. Carbon monoxide adsorbed

on Pt surfaces is a sensitive probe of two properties: the accessibility of platinum surface

area and the electronic structure of the platinum surface [18]. The intensity of CO bands

reflects the number of adsorbed CO molecules, while the CO frequency reflects the

electron donating capabilities of the surface platinum atoms. The reduction of surface

platinum atoms affects the interaction with adjacent pyrrolidone rings and chemisorbed

carbon monoxide differently (Scheme 1). The reduction of Pt nanoparticles decreases

charge-transfer interactions between carbonyl groups of pyrrolidone rings and platinum

as followed using SERS [5], but increases the strength of CO chemisorption as a

consequence of increased electron back-donation from Pt atoms to CO. We can only

speculate on the reason for the maximum in the graph in Figure 8. The frequency of

adsorbed carbon monoxide is a monitor of subtle changes in the electronic state of

platinum atoms during thermal decay of capping agent in H2 atmosphere. At low heating

temperatures (80 – 160 °C), the donation of electron lone pairs from amide carbonyls

bonded to Pt δδδδ+ atoms decreases as surface platinum atoms in the reduced state repel the

lone pair electrons of a carbonyl group, creating a new site on metallic nanoparticles for

CO adsorption (Scheme 1). Above 160 °C, the reduction of the Pt surface plays a

dominant role in lowering the CO stretch frequency. Shielding of the metal surface by

products of the PVP thermal decomposition also takes place. We concluded, that the Pt

14

and Rh atoms on the surface of nanoparticles with ligated pyrrolidone rings and carbon

monoxide behave similarly to coordination compounds [19].

CO adsorption on Rh-PVP and Pt-PVP after RedOx treatment.

The dependence of CO band intensity vs. reduction temperature of Pt-PVP and Rh-PVP

in an excess of gas-phase carbon monoxide was studied. The samples were first subjected

to reduction for 10 min in H2 flow at temperatures ranging from 50 to 300 °C. Then the

samples were pumped down to ~10-4 torr at room temperature and then repressurized

with 260 ± 5 torr of CO and the FTIR spectrum measured. Thereafter oxygen was

flowed at 30 °C over the sample to remove adsorbed carbon monoxide and this procedure

was repeated with the same sample for treatments at 50, 100, 150, 200, 250 and 300 °C.

It was found that about 25 min was necessary to reach saturation of the CO band

intensity. Figure 9a shows that for Rh-PVP + CO, the shape of carbon monoxide bands

transforms with time . Initially there are two CO bands, one at 2032 cm-1 (atop-linear

structure) and the other at 1830 cm-1 (threefold hollow adsorption); after 5 min, a new

band appears at 1880 cm-1 (bridge adsorption). For comparison, all spectra in Figure 9a

were normalized to the absorbance of the 2032 cm-1 band. The simultaneous appearance

of the two bands (2032 cm-1 and 1830 cm-1) demonstrates that adsorption energies for

atop sites and for hollow sites are very close and this is in good agreement with results of

theoretical calculations of adsorption energies for CO on Rh(111) surfaces [20].

Increasing the temperature from 25 to 300 °C in H2 flow resulted in decreasing both CO

stretching frequency (Figure 9b) and integrated intensity (Figure 9c). The decrease of

carbon monoxide frequency chemisorbed on the surface of nanoparticles is due to

reduction of the surface of Rh and Pt particles, thereby increasing the dπ→π* back-

15

bonding to CO. The CO bands after treatment in H2 above 300 °C can no longer be

observed, suggesting that the residual products are occupying surface sites that are

favorable for CO bonding. It should be noted that FTIR data for CO adsorption via in-situ

measurements (Figure 7c, 8a and 8b) and in excess pressure of ‘external’ CO at room

temperature (Figure 9) are distinctive and unclear. The first one monitors the in-situ

transformation of the nanoparticle interface with Pt in highly reduced state with the CO

frequencies ranging from 2020 to 1950 cm-1. In RedOx mode the surface of Pt-PVP is

more oxidized, as shown by the range of 2060-2040 cm-1 for the CO frequencies.

Polyamide-polyene products. Degradation of Pt-PVP and Rh-PVP in an

oxygen atmosphere.

More complex changes of the PVP with increasing temperature were observed for neat

PVP and Rh-PVP and Pt-PVP in O2. While the intensities of all bands for Pt-PVP and

Rh-PVP upon thermal decomposition gradually decreased, several new absorption bands

of residual products appeared.

The data for Rh-PVP is summarized in Figure 10. Capping PVP is more stable in

hydrogen than in oxygen (Fig. 10 a, b). In reducing conditions Rh-PVP loses carbonyl

and CH groups starting at 220 °C and the ratio of integral intensities >C=O/CH increases

(Figure 10 a, c). As can see in Figure 10c, the spectra of initial product at 30 °C and those

of the residual at 220 °C are very close, which means that the capping PVP (55 K, MW)

releases low molecular weight fragments. Noticeable decomposition of capping PVP in

oxygen starts at 170 °C and ratio >C=O/CH has a maxima near 260 °C, which reflects

appearance new of oxygenates with carbonyl bands at 1718 and1780 cm-1 (Fig. 10 b, d).

The decrease of the aliphatic CH band intensity relative to that of the >C=O band of the

16

pyrrolidone rings demonstrates that during heating of Rh-PVP, the aliphatic units of the

backbone undergo thermal degradation, probably due to dehydrogenation (Scheme 2).

Scheme 2

Additional evidence in favor backbone dehydrogenation may be seen in the spectra of

Rh-PVP at temperatures above 180 °C (Figure 11). An absorbance at 1590-1610 cm-1

appears and shifts to lower frequencies with increasing temperature. This band may be

assigned to the C=C stretch of polyene sequences [17] (Figure 11 a, b). The band at 1610

cm-1 also appeared in oxidizing environments for pure PVP, Rh-PVP, and Pt-PVP, but at

different temperatures. In Fig. 11c one can see the appearance new bands at 3340, 3080,

2230, 2157 and 1590 cm-1 at high temperature. IR peak assignments is presented in Table

2 for residual products formed at 280 °C (ramp 3.2 °C /min) from pure PVP and Rh-PVP

in O2 flow. As seen in Fig. 11 and Table 2, during the thermal decay of pure PVP and of

PVP capped Rh nanoparticles, new compounds form with N-H bonds (3330-3342 cm-1),

nitrile bonds (2161-2254 cm-1), polyene >C=C< groups (3089 and 1600 cm-1, Figure 12a-

b), and possibly aldehyde groups (1703-1735 and 1335 cm-1). The positions of these

adsorption bands are slightly shifted on Rh-PVP/Al compared to PVP/Al and new peaks

appear in case of Rh-PVP as well.

17

TABLE 2

PVP in O2 Rh-PVP in O2 Assignment

______________________________________________________

3330 3342 N—H

3082 3089 C—H, vinyl

2968 C-H, νas (CH3)

2940 2937 C—H, νas(CH2)

2254 2246 ν(C≡N)

2226 2222 ν(C≡N)

2161 ν(C≡N)

1780 1780 >C=O or C≡N

1735-1724 1718-1703 ν(>C=O), amide I

1600 1595 ν(C=C—C=C) ν(arom. ring, amide II)

1328 1335 δ(CH), -CHO,

1158 ν(C—C), CCHO

It is believed that the differences between the spectra of pure PVP and those of Rh-PVP

residual products originates from active oxygen attack (O2 spill-over from Rh surface) of

Rh-PVP, and from the fact that drop-casted layers of Rh-PVP after high-temperature

treatment in O2, transform into a thin layer of residual products bonded to the surface of

rhodium nanoparticles. The residual products of deposited on Al-foil PVP degradation

18

bond to the surface of Al2O3. The FTIR spectra of the residuals also reveal some

absorption bands similar to those that appear in the spectra of residual non-volatile char

formed during the process of thermal degradation of polyamide [21]. PVP is an addition

polymer that has aliphatic hydrogen atoms along the backbone that are easily transferred

following homolytic bonds cleavage in the temperature range of 200°C -300°C [22]. One

may speculate that above 250°C, capping PVP transforms into a polyamide-polyene

structure.

We measured the deep UV-Raman spectrum of the non-volatile char that formed from

Rh-PVP and Pt-PVP deposited on Al-foil after oxidation in O2 and after reduction in H2

(heating to 300 °C, 3.2 °C/min). It was found that in the Raman spectra of both chars (Rh

and Pt nanoparticles) intense and broad peaks appear at ~1595 cm-1 and ~1380 cm-1

(Figure 12c), indicative of the presence of disordered graphite . These two peaks may be

assigned to the G and D vibrations (so-called carbon sp2-type) of disordered amorphous

carbon [23]. The weak peak at 1030 (cm-1) (T) may be a trace of sp3-type tetrahedral

amorphous carbon. Recently we have shown by deep UV-Raman that Pt-PVP thermally

degrades in the air at 350 °C to amorphous carbon [6].

Thus the heating of PVP stabilizer capped Rh nanoparticles occurs in a number of steps:

conformational change of the PVP, production of polyene sequences with polyamide-

polyene fragments and several end-groups such as –C≡N and CHO, and finally the

formation of char, which is made up of amorphous carbon. At temperatures above 250

°C in O2, a thin layer of a complex composite material forms on the surface of Rh and Pt

nanoparticles, comprised of polyamide - polyene fragments and amorphous carbon

(Figure 12d).

19

CONCLUSIONS

We have shown that spectroscopic techniques are very effective tool to study the

degradation of poly(N-vinylpyrrolidone) stabilizer of rhodium and platinum

nanoparticles, the electronic structure of the metal/PVP interface and the aggregation of

nanoparticles. The spectroscopic study of the thermal decay of PVP stabilizer in H2 and

O2 atmospheres shows the formation of ‘polyamid-polyene’ like material (above 250

°C), which then transforms into amorphous carbon (above 300 °C). The behavior of low-

coordinated surface Rh and Pt atoms with ligated CO and amid groups of PVP resembles

that of ‘surface coordinated compounds’.

ACKNOWLEDGMENT

We are truly grateful to Prof. Herbert L. Strauss for his valuable advice during

preparation of this paper. This work was supported by the Director, Office of Science,

Office of

Basic Energy Sciences, Division of Chemical Sciences, Geological and

Biosciences and Division of Materials Sciences and Engineering of the

U.S. Department of Energy under contract No. DE-AC02-05CH11231.

20

References

1 Fievet, F.; Lagier, J. P.; Blin, B.; Beaudoin, M.; Figlarz, M. Solid State Ionics 1989,

32/33, 198; Viau, G.; Fievet-Vincent, F; Fievet, F. Solid State Ionics 1996, 84, 259;

Song, H.; Kim, F.; Connor, S.; Somorjai, G. A.; Yang, P. J. Phys. Chem. B. 2005,

109, 188; Wiley, B.; Sun, Y.; Mayers, B.; Xia, Y. Chem. Eur. J. 2005, 11, 454.

2 Didier Astruc, Ed. Nanoparticles and Catalysis, 2007, Wiley-VCH, Weinheim; Song,

H; Rioux, R.M.; Hoefelmeyer, J.D.; Komor, R.; Niez, K.; Grass, M.; Yang, P.D.;

Somorjai G.A. J. Am. Chem. Soc. 2006, 128, 3027; Bratlie, K.M.; Klliewer, C.J.;

Somorjai G.A. J. Phys. Chem. B, 2006, 110, 17925; Somorjai, G.A. and Park, J.Y.

Angew. Chem. Int. Ed. 2008, 47/48, 9212

a) Aliaga, C.; Park,J.Y.; Yamada, Y.; Lee, H.S.; Tsung, C.K.; Yang, P.D.; Somorjai,

G.A. J. Phys. Chem. C, 2009, 113/15, 6150.

3 Zhang, Y; Grass, E.G.; Habas, S.E.; Tao, F; Zhang, T; Yang, P.; Somorjai,G.A.

J. Phys. Chem. C. 2007, 111, 12243.

4 Borodko, Y.; Ager, J.W.; Marti, G.E.; Song, H.; Niesz, K. and Somorjai, G.A.

J. Phys. Chem. B 2005, 109, 17386.

5 Borodko, Y.; Humphrey, S.M.; Tilley, T.D.; Frei, H. and Somorjai, G.A.

J. Phys. Chem. C. 2007, 111, 6288.

6 Borodko, Y.; Habas, S.E.; Koebel, M.; Yang, P.; Frei, H. and Somorjai, G.A. J.

Phys. Chem. B 2006, 110, 23052

7 Zettsu, N.; McLellan, J.M.; Wiley, B.; Yin, Y.; Li, Z.-Y. and Xia, Y. Angew. Chem.

Int. Ed. 2006, 45, 1288. Zhang, Y.; Grass, M.E.; Habas, S.E.; Tao,F.; Zhang, T.;

Yang, P. and Somorjai, G.A. J. Phys. Chem. C 2007, 111, 12243. Humphrey, S.M.;

21

Grass, M.E.; Habas, S.E.; Niez, K.; Somorjai, G.A. and Tilley, T.D. Nano Letters

2007, 7/3, 785

8 Bohren, C.F.; Huffman, D.R. Absorption and Scattering of Light by Small Particles.

Wiley-VCH, Weinheim, 2004.

9 Harrick, N.J. Applied Optics 1971, 10/10, 2344. Hind, A.R. and Chomette, L. The

determination of thin film thickness using reflectance spectroscopy. Varian UV at

Work No. 090, www.varianinc.com. Guan, Y.; Yang, S.; Zhang, Y., Xu, J., Han,

C.C.; Kotov, N.A. J. Phys. Chem. B 2006,110/27, 13484. Williams, C.T.; Chen,

E.K.-Y. ; Takoudis, C.G. and Weaver, M.J. J. Phys. Chem. B 1998, 102, 4785.

10 Grass, M.E.; Zhang, Y.; Butcher, D.R.; Park,J.Y.; Li,Y.; Bluhm, H.; Bratlie K.M.;

Zhang, T. and Somorjai, G.A. Angew. Chem. Int. Ed. 2008, 47, 8893. Munoz, A.;

Munuera, G.; Malet, P.; Gonzalez-Elipe, A.R. and Espinos, J.P. Surf. and Interface

Analysis 1988, 12, 247.

11 McCabe, R.W.; Wong, C.; Woo, H.S. J. Catal. 1988, 114, 354. Huzinga, T.; Van

Grondelle, J.; Prins, R. Appl. Catal. 1984, 10, 199.

12 Weaver, J.H. Physical Review B 1975, 11/4, 1416. Weaver, J.H.; Olson, C.G.;

Lynch,D.W. Phyical. Review B 1977, 15/8, 4115. Pierce, D.T. and Spicer, W.E.

Phys. Stat. Sol. (b) 1973, 60, 689.

13 Anno, E. Physical Review B 1996, 53/7, 4170. Anno, E.; Tanimoto, M. J. Appl.

Phys., 2000, 88/6, 3426.

14 Langhammer, C.; Yuan, Z.; Zoric, I. and Kasemo, B. Nano Letters, 2006, 6/4, 833.

15 a) Kreibig, U. and Fragstein, C. Z. Phys. 1969, 224, 307; b) Lin,S.; Li, M.;

Dujardin, E.; Girard, C.; Mann, S. Adv. Mater. 2005, 15, 2553

22

16 Yoshida, M. and Prasad, P.N. Appl. Optics 1996, 35/9, 1500. Scheirs, J.; Bigger,

S.W.;Then, E.T.H.; Billingham, N.C. J. Polym. Sci. B: Polymer Physics, 1993, 31,

287. Peniche, C.; Zaldivar, D.; Pazos, M.; Paz, S.; Bulay, A.; Roman, J.S J. Appl.

Polym. Sci., 1993, 50, 485. Du, Y.K.; Yang,P.; Mou, Z.G.; Hua, N.P.; Jiang, L. J.

Appl. Polymer Science 2006, 99, 23.

17 Colthup, N.B.; Daly, L.H.; Wiberley, S.E. Introduction to Infrared and Raman

Spectroscopy, Acad. Press, 1990 San Diego.

18 Thomas, J.M. and Thomas W.J. Introduction to the Principles of Heterogeneous

Catalysis, Academic, London, 1967. Knozinger, H, in: Ertl, G.; Knozinger, H.;

Weitkamp, J. (Eds.), Handbook of Heterogeneous Catalysis, vol. 2, Wiley-VCH,

Weinheim, 1997, p. 707. Davydov, A. Molecular Spectroscopy of Oxide Catalyst

Surfaces, Willey, Chichester, 2003.

19 Hemminger, J.C.; Muetterties, E.L. and Somorjai, G.A. J. Am. Chem. Soc. 1979, 101,

62. Barth, J.V. Surf. Science, in press 2009 doi:10,1016/j.susc.2008.09.049

20 Mason, S.E.; Grinberg, I. and Rappe, M. Phys. Rev. B, 2004, 69, 161401 (R); Abild-

Pedersen, F.; Andersson, M.P. Surf. Sci., 2007, 601, 1747.

21 Holland, B.J.; Hay, J.N. Polym. Int. 2000, 49, 943. Nyden, M.R. Appl. Spectr. 1999,

53, 1653.

22 Montaudo, G.,Puglisi, C. in Development in Polymer Degradation. Grassie, N. Ed.

Appl. Sci. 1987, 7, London; in Comprehensive Polymer Sci. First Suppl. Pergamon

Press, Oxford, 1992, p. 227.

23

23 Ferrari, A.C.; Robertson, J. Phys. Rev. B 2001, 64/7, 75414; Phil. Trans. R. Soc.

Lond. A 2004, 362/1824, 2477. Hoffmann, G.G.; de With, G.; Loos, J. Macromol.

Symp. 2008, 265, 1-11

FUGURE CAPTIONS

Figure 1. TEM analysis of PVP stabilized Pt nanoparticles (a) and Rh nanoparticles (b).

Figure 2. UV-VIS spectra: A) Absorbance of pure PVP layers (1.3µ, 3.8µ and 4.2µ)

deposited on quartz window; B) Comparison of UV-VIS spectra of Rh-PVP:

absorbance in ethanol solution (a), absorbance of dry layer on quartz window

(b) and absorptance (reflection mode) layer deposited on Al-foil (B, c). C)

Effect of Rh-PVP layer thickness (1.3µ d; 3.8µ e; and 4.2µ f) on the ratio of

intensities: PVP (n → π*) charge-transfer bands Rh ← O=C< ; D)

Absorbance spectra of Rh-PVP for 3µ layer (1) and 5µ layer (2) (transmission

mode). In insert (D, 1a): interference fringe pattern for 3µ layer.

Figure 3. Scheme of possible types for absorption bands Pt- and Rh-PVP in UV-VIS

region. A) Effect of reduction in H2 on CT bands Pt-PVP and Rh-PVP (in-

situ) and B) effect of Rh-PVP oxidation on absorptance spectra (temperature

range 30 -550 °C, ex-situ).

Figure 4. The SEM and XRD data for Rh-PVP (A) and Pt-PVP (B) heated in air.

Figure 5 The scheme of reflectance measurements: A) with reflecting media behind the

sample, and reference; B) without reflectors behind the sample and reference.

C) Reflectance spectra of Rh-PVP after heating layer of Rh-PVP deposited on

quartz at 550°C and 350 °C (Scheme B).

24

Figure 6. a) Evaluation of the Rh aggregates size after heating in air Rh-PVP

nanoparticles (7 nm) based on optical plasmon-resonance location taken from

the work of Anno [13] and b) comparison effective size of Rh aggregates with

surface plasma-resonance (SPR) at 240 nm with XRD measurements.

Figure 7. Temperature dependence of FTIR-DRIFT spectra of PVP in H2 flow. The

heating only slightly affect the CH stretching bands (a), however CH2

deformation modes: scissors (1430 cm-1) and wagging (1300 cm-1) are more

sensitive to the temperature increase (b). c) FTIR in-situ spectra of Pt-PVP

heated in flowing H2. During degradation, elimination of CO from side

pyrrolidone rings may occur. Carbon monoxide adsorbed on Pt nanoparticles

appeared in the temperature range of 100–200 °C.

Figure 8. The effect of heating Pt-PVP nanoparticles in H2: on the adsorbed carbon

monoxide frequencies: a) linear atop (n) and bridging (����) structures; and on

the intensities of these bands (b) and carbonyl band of capping PVP ( �)

Figure 9. a) The change of relative intensities of the three CO bands: 2032 cm-1 (atop

structure), 1880 cm-1 (bridging structure) and 1830 cm-1 (threefold hollow

structure) with time for (Rh-PVP) + CO. The CO stretch absorbance for atop

structure was used to normalize intensities. b) The effect of heating Pt-PVP in

hydrogen on CO bands position (■- linear atop structure; □-bridging

structure) and (c) CO bands intensities.

Figure 10. FTIR spectra of Rh-PVP in H2 and O2 media. a) The ratio of integral

intensities >C=O/CH during heating in H2 (3.2 °C /min); b) The ratio of

integral intensities >C=O/CH during heating in O2 (3.2 °C /min); c, d) The

25

overview in-situ spectra of Rh-PVP in the range 1000-3500 cm-1 and

temperature range 35-250 °C in H2 ( c) and O2 (d).

Figure 11. Effect of heating Rh-PVP nanoparticles in O2 flow on FTIR-DRIFT spectra in

the temperature range 30 - 300°C. a) Overview FTIR spectra of Rh-PVP in

the spectral range 3600– 850 cm-1. b) Appearance at high temperature a new

bands at 1775, 1714 (>C=O) and at 1610 cm-1 (>C=C<). c) Comparison the

initial spectrum of Rh-PVP at 50 °C (black line) with spectrum of residual at

285 °C (grey line). Both spectra were normalized to the carbonyl absorption.

The residual has new bands at 3340 (N-H), 3080, 1590 (H-C=C<), 2230 and

2157 cm-1 (-C≡N), which are characteristic of ‘polyamid-polyene’ like

structure.

Figure 12. a, b) Appearance and gradual increase intensity of band at 1610 cm-1 (carbon-

carbon double bond) and absorbance above 1700 cm-1 (new carbonyl

bonds). c) Raman spectrum of residual after heating Rh-PVP at 350 °C in

H2. d) The cartoon shows the intermediate formation of residual product

with dehydrated backbone and some end groups.

26

Figure 1.

TEM analysis of PVP stabilized Pt nanoparticles (a) and Rh nanoparticles (b).

27

Figure 2. UV-VIS spectra: A) Absorbance of pure PVP layers (1.3µ, 3.8µ and 4.2µ)

deposited on quartz window; B) Comparison of UV-VIS spectra of Rh-PVP:

absorbance in ethanol solution (a); absorbance of dry layer on quartz window

(b) and absorptance (reflection mode) layer deposited on Al-foil (B, c). C)

Effect of Rh-PVP layer thickness (1.3µ, d; 3.8µ, e; and 4.2µ, f) on the ratio of

intensities: PVP (n → π*) charge-transfer bands Rh ← O=C< ; D)

Absorbance spectra of Rh-PVP for 3µ layer (1) and 5µ layer (2) (transmission

mode). In insert (D, 1a): interference fringe pattern for 3µ layer.

.2

.4

.6

200 400 600 800

a

b

c

nm

Absorbance

B

0

20

40

200 400 600 800

d

e

f

nm

c

(100

-T-R

)%

0

.1

.2

.3

200 400 600 800

nm

1

2 Absorbance

-.004

0

.004

500 600 700 800 900 nm

1a D

2

1

0

.5

1

200 400 600 800 nm

Absorbance

A

2

1

3

N

CCH2

O

H

n

28

10

20

30

40

50

200 300 400 500 600 700 nm

(100-T-R)%

30°C

200°C 400°C

550°C

Rh-PVP

20

40

60

200 300 400 500 600 700

Rh-PVP

Pt-PVP

30°C

30°C

170°C

170°C

nm

(100-T-R)%

A

B

1. Band of non - bonded amide

group of PVP (n ���� ππππ *)

2. CT band >C=O --- Pt (n ���� d*)

3.

Interband transition 4

Surface plasmon resonance Metal

nanopar ticle

Stabilizer

Figure 3 Scheme of possible types for absorption bands Pt- and Rh-PVP in UV-VIS

region (top). (A) Effect of reduction in H2 on CT bands Pt-PVP and Rh-PVP

(in-situ) and (B) effect of Rh-PVP oxidation on absorptance spectra

(temperature range 30 -550 °C, ex-situ measurement).

29

Figure 4 The SEM and XRD data for Rh-PVP (A) and Pt-PVP (B) heated in air.

30

Figure 5

Figure 5 The scheme of reflectance measurements: A) with reflecting media behind the

sample, and reference; B) without reflectors behind the sample and reference.

C) Reflectance spectra of Rh-PVP after heating layer of Rh-PVP deposited on

quartz at 550 °C and 350 °C in air (Scheme B).

Ir

Is

Ir

Is S S

R R

A B

50

60

70

80

90

200 300 400 500 600 700

244 nm

nm

Reflectance, a.u.

a.u.

--------------------------

350 °°°°C

550 °°°°C

C

31

Figure 6 a) Evaluation of the Rh aggregates size after heating in air Rh-PVP

nanoparticles (7 nm) based on optical plasma-resonance location taken from the

work of Anno [13] and b) comparison effective size of Rh-clusters with surface

plasma-resonance at 240 nm with XRD measurements.

30 °°°°C 450 °°°°C

4/3 ππππr3 8/3 ππππr

3 4/3 ππππ(r+∆)3

5.5 nm 11 nm 7 nm XRD

∼∼∼∼30 nm

SPR

0 50

100 150

200 250

300

200

240

280

320

360

P lasma resonance , nm

Diameter of Rh islands, nm

a

b

32

Figure 7. Temperature dependence of FTIR-DRIFT spectra of PVP in H2 flow. a) The

heating only slightly affect the CH stretching bands, however CH2 deformation modes:

scissors (1430 cm-1) and wagging (1300 cm-1) are more sensitive to the temperature

increase (b). c) FTIR in-situ spectra of Pt-PVP heated in flowing H2. During degradation,

elimination of CO from side pyrrolidone rings occurs. Carbon monoxide adsorbed on Pt

nanoparticles appeared in the temperature range of 100 °C – 200 °C.

.1

.2

3000

2900

2800

33 º C

ºC

Absorbance

0

.5

1800 1600

1400 1200

33 º C

225

225

a b

0

.2

.4

2000 1800 1600 1400 1200

Absorbance

300 °°°°C

278 °°°°C

218 °°°°C

35 °°°°C

cm -1

0

.02

.04

2000 1900

.06

1800

116 ºC

156 ºC 198 ºC

c

33

Figure 8. The effect of heating Pt-PVP nanoparticles in H2. a) On the adsorbed carbon

monoxide frequency: linear atop (n) and bridging (����) structures, and on the

intensities of these bands and carbonyl band >C=O of capping PVP ( �).

a

b

34

Scheme 1

NHC

O

CH2

C

O

CT

e

NHC

O

CH2

C

O

C

O

H2

35

Figure 9.

a) The change of relative intensities of the three CO bands: 2032 cm-1 (atop structure),

1880 cm-1 (bridging structure) and 1830 cm-1 (threefold hollow structure) with time for

Rh-PVP + CO. The CO stretch absorbance for atop structure was used to normalize

intensities. b) The effect of heating Pt-PVP in hydrogen on CO bands position (■- linear

atop structure; □-bridging structure) and CO bands intensities (c).

50 100 150 200 250 300 0.0

0.2

0.4

0.6

0.8

1.0

50 100 150 200 250 300

1980

2000

2020

2040

2060

Absorbance, normalized

Temperature, °C

ν

ν

ν

ν

, cm

-1

c

0

.02

.04

.06

2200 2100 2000 1900 1800

1830 1880

2032

Absorbance, normalized

cm-1

min 1

3

5

14

24

50

a

b

36

Figure 10. FTIR spectra of Rh-PVP in H2 and O2 media. a) Normalized integral

intensities of >C=O and CH. Heating in H2 (3.2 °C /min); b) Normalized integral

intensities >C=O and CH during heating in O2 (3.2 °C /min); The overview in-situ

FTIR spectra of Rh-PVP in the range 1000-3500 cm-1 and temperature range 35-250 °C

in H2 ( c) and O2 (d).

0

.05

.1

.15

3500 3000 2500 2000 1500 1000

Absorbance

Rh - PVP in H 2

0

.2

.4

.6

3500 3000 2500 2000 1500 1000

Rh-PVP in O2

cm-1

Absorbance

50 100 150 200 250 3000.0

0.2

0.4

0.6

0.8

1.0

1.2

a

C-H

>C=OIntensity, normalized

Temperature, oC

50 100 150 200 250 3000.0

0.2

0.4

0.6

0.8

1.0

1.2 b

C-H

>C=O

Temperature,oC

a

c

d 35 ºC 140 ºC 220 ºC 250 ºC

37

Figure 11.

Effect of heating Rh-PVP nanoparticles in O2 flow

on FTIR-DRIFT spectra in the temperature range

30 - 300°C. a) Overview FTIR spectra of Rh-PVP

in the spectral range 3600– 850 cm-1.

b) Appearance at high temperature a new bands

at 1775, 1714 (>C=O) and at 1610 cm-1 (>C=C<).

c) Comparison the initial spectrum of Rh-PVP at 50

°C (black line) with spectrum of residual at 285 °C

(grey line). Both spectra were normalized to

the carbonyl absorbance. The residual has new

bands at 3340 (N-H), 3080, 1590 (H-C=C<),

2230 and 2157 cm-1 (-C≡N), which are

characteristic of ‘polyamid-polyene’ like structure.

0

.5

1

3500 3000

2500

2000

1500

1000

30 ° C 200 ° C 251 ° C 268 ° C

300 ° C 285 ° C

0

.4

.8

1800 1600 1400 1200 1000

31 ° C

200 ° C

268 ° C

285 ° C

302 ° C

1610

1690

1775 .............

.

........

.

Absorbance

0

.1

.2

3500 3000 2500 2000 1500 1000

3340

3080

2230

2157

1714 1680

1590

cm-1

a

b

c

38

Figure 12.

a, b) Appearance and gradual increase intensity of band at 1610 cm-1 (carbon-

carbon double bond) and absorbance above 1700 cm-1 (new carbonyl bonds). c)

UV-Raman spectrum of residual after heating Rh-PVP at 350 °C in H2. d) Cartoon

shows the intermediate formation of residual product with dehydrated backbone and

new end groups.

2

1 2 3 4 5

0

.2

.4

.6

.8

1

1800 1750 1700 1650 1600 1550

0

.2

.4

.6

.8

1

1500 1700 1600

>C=C <

1610

1692 >C=O ABSORBANCE

30 ° C

183 ° C

200°C

218 ° C 251 ° C

.2

.3

.4

1640 1600 1560 cm-1

...............................................................

.......

1610 cm - 1

251 ° C

218 ° C

200 ° C 183 ° C

0

5

10

1800 1600 1400 1200 1000

1600 G

1378 D

1030

cm-1

Intensity, a.u.

c b a

30 – 200°C 200 – 300°C 250 – 550°C

NC

CH

O

CH2

H

N C

C H

O

N C

O

C N

RhRhRhRh

d

39

Table of Contents (TOF) Image.

30 – 200°C 200 – 300°C 250 – 550°C

NC

CH

O

CH2

H

N C

C H

O

N C

O

C N

RhRhRhRh